Abstract: Systems, methods, and computer-readable storage media for determining user engagement levels during a presentation of content. The system first collects data associated with a user session at a client device. Next, the system predicts a user engagement level during the user session by applying an engagement predicting rule to the data. The system can predicts respective user engagement levels for various segments of the presentation by applying one or more engagement predicting rules to the data. The system then presents invitational content based on the user engagement level.

Abstract: Systems, methods, and computer-readable storage media for dynamic radio content. The system generates a media item to be inserted into a media stream. The media item can include multiple segments associated with a set of discrete components. Next, the system inserts the media item into the media stream. In response to a triggering event, the system then re-generates the media item by assembling the multiple segments using a new set of the discrete components. The system then inserts the re-generated media item into the media stream.

Abstract: Systems, methods, and computer-readable storage media for determining user engagement levels during a presentation of content. The system first collects data associated with a user session at a client device. Next, the system predicts a user engagement level during the user session by applying an engagement predicting rule to the data. The system can predicts respective user engagement levels for various segments of the presentation by applying one or more engagement predicting rules to the data. The system then presents invitational content based on the user engagement level.

Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for presenting event data in a heat-map interface. Event data describing user interactions with a content item can be gathered and used to generate the heat map. The gathered event data can be filtered prior to processing to exclude unnecessary or irrelevant event data. The event data can be filtered based on numerous criteria such as a determined type of user interaction, demographic data, location of the event data, content item, etc. Filtering rules can be based on an analysis of the content item to determine what event data to exclude and current performance of the system. The filtered event data can be presented in a heat-map interface that represents the frequency at which event data occurs on an item of content based on location. The event data is analyzed to determine optimization rules that dictate selection of content items.

Abstract: Systems, methods, and computer-readable storage media for determining user engagement levels during a presentation of content. The system first collects data associated with a user session at a client device. Next, the system predicts a user engagement level during the user session by applying an engagement predicting rule to the data. The system can predicts respective user engagement levels for various segments of the presentation by applying one or more engagement predicting rules to the data. The system then presents invitational content based on the user engagement level.

Abstract: Systems, methods, and computer-readable storage media for determining user journeys during presentations of content. The system first determines an average time spent for sessions associated with a presentation of content. Next, the system identifies a representative group of sessions from the sessions by identifying each of the sessions having a respective time spent within a statistical range from the average time spent for the sessions. The system then determines a most common path of events from the representative group of sessions to yield a most common user journey associated with the presentation of content.

Abstract: The disclosed technology addresses the need in the art for determining whether a selection of an invitational content item was accidental or intentional based on selection signals that describe the selection. For example, a selection signal can be a selection point, response time, multiple selection count, context of the selection, user preference data, etc. A response to the selection can be based on whether the selection was determined to be accidental or intentional. If the selection was determined to be intentional, the response can be the regular response associated with selection of the invitational content item, i.e. registering the selection for billing and presenting secondary content to the user. If the selection was determined to be accidental, the selection can be ignored. Thus, the selection would not register for billing purposes and secondary content would not be presented to the user.

Abstract: Systems, methods, and computer-readable storage media for dynamic radio content. The system generates a media item to be inserted into a media stream. The media item can include multiple segments associated with a set of discrete components. Next, the system inserts the media item into the media stream. In response to a triggering event, the system then re-generates the media item by assembling the multiple segments using a new set of the discrete components. The system then inserts the re-generated media item into the media stream.

Abstract: Disclosed are systems, methods, and non-transitory computer-readable storage media for an online affiliate program for offline retail businesses. An online content retailer that sells content items (e.g., music, movies, books, etc.) can provide an affiliate program to offline retail businesses (e.g., brick and mortar businesses) whereby, upon entering into a contractual relationship to be an affiliate of the online content retailer, an offline retail business can receive a commission fee for business driven to the online content retailer by the offline retail business (e.g., purchases of content items from the online content retailer driven by the affiliate).

Abstract: Systems, methods, and computer-readable storage media for determining user journeys during presentations of content. The system first determines an average time spent for sessions associated with a presentation of content. Next, the system identifies a representative group of sessions from the sessions by identifying each of the sessions having a respective time spent within a statistical range from the average time spent for the sessions. The system then determines a most common path of events from the representative group of sessions to yield a most common user journey associated with the presentation of content.

Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for managing a viral invitational content campaign. Invitational content is delivered with a user interface element that, upon selection, transmits a forwarding request to the content delivery server. The content delivery server generates a link with a unique identifier and creates a forwarding entry that includes the unique identifier and identifies the item of invitational content. The link is returned in response to the forwarding request. Upon selection of the link, a request including the unique identifier is transmitted to the content delivery server which accesses the corresponding forwarding entry to return the corresponding item of invitational content. A viral campaign dictates a target spend range and throttling methods are utilized to manage the spend. Invitational content forwarding, delivery cost, and content selection are variable to manage the spend. Calculated metrics are also used to manage spend.

Abstract: Systems, methods, and computer-readable storage media for determining user engagement levels during a presentation of content. The system first collects data associated with a user session at a client device. Next, the system predicts a user engagement level during the user session by applying an engagement predicting rule to the data. The system can predicts respective user engagement levels for various segments of the presentation by applying one or more engagement predicting rules to the data. The system then presents invitational content based on the user engagement level.

Abstract: The disclosed technology addresses the need in the art for determining whether a selection of an invitational content item was accidental or intentional based on selection signals that describe the selection. For example, a selection signal can be a selection point, response time, multiple selection count, context of the selection, user preference data, etc. A response to the selection can be based on whether the selection was determined to be accidental or intentional. If the selection was determined to be intentional, the response can be the regular response associated with selection of the invitational content item, i.e. registering the selection for billing and presenting secondary content to the user. If the selection was determined to be accidental, the selection can be ignored. Thus, the selection would not register for billing purposes and secondary content would not be presented to the user.

Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for presenting event data in a heat-map interface. Event data describing user interactions with a content item can be gathered and used to generate the heat map. The gathered event data can be filtered prior to processing to exclude unnecessary or irrelevant event data. The event data can be filtered based on numerous criteria such as a determined type of user interaction, demographic data, location of the event data, content item, etc. Filtering rules can be based on an analysis of the content item to determine what event data to exclude and current performance of the system. The filtered event data can be presented in a heat-map interface that represents the frequency at which event data occurs on an item of content based on location. The event data is analyzed to determine optimization rules that dictate selection of content items.

Abstract: The present technology delivers a multi-user targeted content package to a group of users. The multi-user targeted content package can include single-user and multi-user interactive content. A multi-user targeted content package can be delivered in bursts so that multiple users receive the content package during the same time period. In some cases, users can be grouped into instances of the multi-user targeted content package based on factors such as location, language, demographics, bandwidth, skill level, and/or other similarity measures. Additionally, some multi-user content can require a minimum number of users before the content becomes fully active, such as a multi-player game.